633559 - Sporadic reftest/tests/layout/reftests/font-matching/weightmapping-*.html | assertion count 3 is more than expected 0 assertions, with "Potential deadlock between Lock@xxx and Lock@yyy"

Status: RESOLVED WORKSFORME

(Core :: XPCOM, defect)

Description

[Daniel Holbert [:dholbert]]

Attachment: snippet of failure log

http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1297445126.1297448601.2494.gz
Rev3 WINNT 6.1 mozilla-central debug test reftest on 2011/02/11 09:25:26
s: talos-r3-w7-009

REFTEST TEST-UNEXPECTED-FAIL | file:///c:/talos-slave/test/build/reftest/tests/layout/reftests/font-matching/weightmapping-12579.html | assertion count 3 is more than expected 0 assertions

The assertions are all like:
###!!! ASSERTION: Potential deadlock between Lock@d7638c0 and Lock@d77de30: 'Error', file e:/builds/moz2_slave/cen-w32-dbg/build/obj-firefox/xpcom/build/nsAutoLock.cpp, line 318

Posting larger snippet of log as attachment.

This looks basically identical to bug 594666, except that bug has always been limited to jsreftest -- this is the first time I've seen this affecting "proper" reftests.

Comment 1

[Daniel Holbert [:dholbert]]

Philor points out that he's reported two earlier instances of this happening in reftests, on bug 594666:

bug 594666 comment 59:
> http://tinderbox.mozilla.org/showlog.cgi?log=TraceMonkey/1294257559.1294260563.2849.gz
> Rev3 WINNT 6.1 tracemonkey debug test reftest on 2011/01/05 11:59:19
> s: talos-r3-w7-033
> 
> REFTEST TEST-UNEXPECTED-FAIL |
> file:///c:/talos-slave/test/build/reftest/tests/layout/reftests/font-matching/weightmapping-12579.html
> | assertion count 3 is more than expected 0 assertions

Bug 594666 comment 175:
> http://tinderbox.mozilla.org/showlog.cgi?log=Firefox/1296532508.1296536670.25585.gz&fulltext=1
> Rev3 WINNT 6.1 mozilla-central debug test reftest on 2011/01/31 19:55:08
> s: talos-r3-w7-017
> 
> REFTEST TEST-UNEXPECTED-FAIL |
> file:///c:/talos-slave/test/build/reftest/tests/layout/reftests/font-matching/weightmapping-7.html
> | assertion count 3 is more than expected 0 assertions

I'm keeping this bug separate from bug 594666 for now, so we can track these (relatively-infrequent) reftest failures separate from the (very frequent) jsreftest failures, in case that ends up being a useful/relevant distinction.

Comment 2

[Ed Morley [:emorley]]

Mass marking whiteboard:[orange] bugs WFM (to clean up TBPL bug suggestions) that:
* Haven't changed in > 6months
* Whose whiteboard contains none of the strings: {disabled,marked,random,fuzzy,todo,fails,failing,annotated,leave open,time-bomb}
* Passed a (quick) manual inspection of bug summary/whiteboard to ensure they weren't a false positive.

I've also gone through and searched for cases where the whiteboard wasn't labelled correctly after test disabling, by using attachment description & basic comment searches. However if the test for which this bug was about has in fact been disabled/annotated/..., please accept my apologies & reopen/mark the whiteboard appropriately so this doesn't get re-closed in the future (and please ping me via IRC or email so I can try to tweak the saved searches to avoid more edge cases).

Sorry for the spam! Filter on: #FFA500